The present invention relates to the magnetic resonance imaging art. It finds particular application in conjunction with insertable gradient coils for high speed imaging techniques and will be described with particular reference thereto.
Magnetic resonance imagers commonly include a large diameter, whole body gradient coil which surrounds a patient receiving bore. Main field magnets, either superconducting or resistive, and radio frequency transmission/reception coils also surround the bore. Although the whole body gradient coils produce excellent linear magnetic field gradients, they have several drawbacks. With large diameter gradient coils, the slew rate is sufficiently slow that it is a limiting factor on the rate at which gradient magnetic fields can be induced and changed. Large diameter whole body gradient coils are too slow for some of the highest speed magnetic resonance imaging techniques. The energy stored in gradient coils is generally proportional to greater than the fifth power of the radius. Hence, large diameter, whole body coils require large amounts of energy. Further, superconducting main magnets have cold shields disposed around the bore. The larger the diameter of the gradient coil, the closer it is to the cold shields and hence the more apt it is to produce eddy currents. More shielding is needed to prevent the whole body gradient coils from inducing eddy currents in the cold shields than would be necessary for smaller diameter coils.
Due to these and other limitations in whole body coils, numerous insertable coils have been developed which are small enough to fit within the bore with the patient. Typically, the insertable coils are customized to a specific region of the body, such as a head coil. Traditionally, head coils have been a cylinder sized to accommodate the human head easily, e.g. 28 cm in diameter. Most brain examinations center around the portion of the brain that is substantially in the same plane as the eye sockets. In a symmetric coil, the magnetic isocenter is configured to be disposed in a common plane with the patient's eyes. For a symmetric coil, the magnetic isocenter is at the physical center of the coil.
As a general rule, the longer the cylindrical head coil, the larger the region over which the gradient is linear and the more linear the region is. However, the patient's shoulders are a limiting factor on the length of a symmetric gradient coil. The shoulders limit the isocenter to about 20 cm at the patient end. Thus, symmetric head coils have heretofore been limited to about 40 cm in length.
In order to achieve the beneficial effects of a longer head coil, head coils have been designed in which the magnetic isocenter is offset toward the patient from the physical geometric center of the coil. See, for example, U.S. Pat. No. 5,278,504 of Patrick, et al. or U.S. Pat. No. 5,177,442 of Roemer, et al. Although asymmetric head coils have beneficial effects on the linearity and the size of the linear region, the improvement is not without an offsetting difficulty. In the main magnetic field, the asymmetric gradient coil is subject to mechanical torques from the magnetic field interactions. In order to counteract these torques, the asymmetric head coils are mounted with rigid mechanical constraints. Even with substantial mechanical structures anchored to the main field magnet assembly, the torque still tends to cause at least mechanical vibration and noise. Adding additional windings to produce a counteracting torque commonly increases the electrical power requirements by about 50% or more.
Although conventional head gradient coils include a Maxwell pair for the z-axis or Golay saddle coils for the x or y-axes on the surface of a cylinder, others have proposed coils in which all windings do not lie on the cylinder surface. "Compact Magnet and Gradient System For Breast Imaging", S. Pissanetzky, et al., SMRM 12th Annual Meeting, p. 1304 (1993) illustrates a compact asymmetric cylinder coil bent up radially at a 90.degree. angle at the field producing end of the coil. The coil is designed for breast imaging with the coil pressed up against the chest. "High-Order, Multi-Dimensional Design of Distributed Surface Gradient Coil", Oh et al., SMRM 12th Annual Meeting, p. 310 (1993) attempts to optimize a gradient surface coil using current flows in three dimensions. One of the problems with the Oh surface gradient coil is that it was difficult to control linearity. Further, the coil was difficult to manufacture due to its complicated shape and high current densities.
The present invention provides a new and improved insertable gradient coil which overcomes the above-referenced problems and others.